Hi there,
to illustrate my question, just a few searches and their results:

Searching for "Book" with:
HS: 1 result
Luke: 0 results

Searching for "book" with:
HS: 1 result
Luke: 1 result

(of course, all of the results pointing to the same entry, which i originally put to the DB as "book")

is this the expected behaviour? Is there no possibilty to differ between "book" and "Book" in HS?

(using 3.0.1)

thx

Which analyzer do you use at query time in Luke vs Hibernate Search? The StandardAnalyzer is used by default in Hibernate Search. This analyzer uses the lowercase filter by default. Did you use the analyzer in Luke?
Remember, in most cases you have to use the same analyzer for indexing and searching.

--Hardy

in both cases i use the germananalyzer

Also the GermanAnalyzer uses by default the lowercase filter. In case you want to differentiate between lower and upper case letters you will have to build your own analyzer using for example the @AnalyzerDef framework omitting the LowerCaseFilter. In most cases, however, fulltext search makes more sense if the case is ignored.

--Hardy

ok, thx for your fast reply (as always)

no problem :)
FYI - http://svn.apache.org/viewvc/lucene/jav ... iew=markup shows how the german analyzer is build using the standard tokenizer, followed by lowercase-, stop- and germanstem-filter.

--Hardy